net: devmem: allow rx-page-size > PAGE_SIZE per dmabuf binding

Every devmem dmabuf binding today hands the page_pool PAGE_SIZE niovs. This caps a single RX descriptor at PAGE_SIZE, burning CPU on buffer churn for large flows. Add a bind-time netlink attribute, NETDEV_A_DMABUF_RX_PAGE_SIZE, that lets userspace request a larger niov size. The value must be a power of two >= PAGE_SIZE. The TX path is changed to always pass PAGE_SIZE. Measurements: Setup: kperf in devmem RX/TX cuda mode, 4 flows, 64 MB messages, 60s, dctcp, num-rx-queues=4, dmabuf-rx/tx-size-mb=2048, 10 runs per niov size, mlx5. CPU Util: niov net sirq % net idle % app sys % app idle % ----- ---------------- ---------------- ---------------- ---------------- 4K 62.38 +/- 8.27 33.40 +/- 7.51 54.15 +/- 10.23 43.67 +/- 10.53 16K 58.91 +/- 5.35 35.23 +/- 5.88 41.05 +/- 8.87 56.42 +/- 9.24 32K 64.12 +/- 0.68 31.09 +/- 1.48 44.54 +/- 3.51 52.63 +/- 3.65 64K 54.69 +/- 5.54 39.67 +/- 5.81 35.47 +/- 3.11 61.97 +/- 3.27 RX app sys % drops ~19% from 4K to 64K. Throughput: niov RX dev Gbps RX flow avg Gbps ----- ---------------- ----------------- 4K 300.63 +/- 53.21 75.16 +/- 13.30 16K 321.35 +/- 28.20 80.34 +/- 7.05 32K 347.63 +/- 2.20 86.91 +/- 0.55 64K 332.11 +/- 14.26 83.03 +/- 3.56 Throughput seems to increase, but the stdev is pretty wide so could just be noise. kperf support (not yet merged): 8837577f92 Acked-by: Stanislav Fomichev <sdf@fomichev.me> Reviewed-by: Mina Almasry <almasrymina@google.com> Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org> Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com> Link: https://patch.msgid.link/20260805-tcpdm-large-niovs-v8-1-3e0225e2808c@meta.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2026-08-27 16:28:58 -04:00 · 2026-08-05 13:42:43 -07:00
parent 55d20f50a2
commit b27a8560ee
8 changed files with 78 additions and 23 deletions
--- a/Documentation/netlink/specs/netdev.yaml
+++ b/Documentation/netlink/specs/netdev.yaml
@@ -6,6 +6,14 @@ doc: >-
  netdev configuration over generic netlink.

 definitions:
+  -
+    type: const
+    name: page-size
+    # Dummy value, codegen needs a number. The real value comes from
+    # the PAGE_SIZE macro in the header below.
+    value: 0
+    header: asm/page.h
+    scope: kernel
  -
    type: flags
    name: xdp-act
@@ -598,6 +606,16 @@ attribute-sets:
        type: u32
        checks:
          min: 1
+      -
+        name: rx-page-size
+        doc: |
+          Size in bytes of each device page the NIC writes into from the bound
+          dmabuf. Must be a power of two and >= PAGE_SIZE; defaults to
+          PAGE_SIZE.
+        type: u32
+        checks:
+          min: page-size
+          max: u32-max

 operations:
  list:
@@ -812,6 +830,7 @@ operations:
            - ifindex
            - fd
            - queues
+            - rx-page-size
        reply:
          attributes:
            - id
--- a/include/uapi/linux/netdev.h
+++ b/include/uapi/linux/netdev.h
@@ -219,6 +219,7 @@ enum {
 	NETDEV_A_DMABUF_QUEUES,
 	NETDEV_A_DMABUF_FD,
 	NETDEV_A_DMABUF_ID,
+	NETDEV_A_DMABUF_RX_PAGE_SIZE,

 	__NETDEV_A_DMABUF_MAX,
 	NETDEV_A_DMABUF_MAX = (__NETDEV_A_DMABUF_MAX - 1)
--- a/net/core/devmem.c
+++ b/net/core/devmem.c
@@ -46,7 +46,7 @@ static dma_addr_t net_devmem_get_dma_addr(const struct net_iov *niov)

 	owner = net_devmem_iov_to_chunk_owner(niov);
 	return owner->base_dma_addr +
-	       ((dma_addr_t)net_iov_idx(niov) << PAGE_SHIFT);
+	       ((dma_addr_t)net_iov_idx(niov) << owner->binding->niov_shift);
 }

 static void net_devmem_dmabuf_binding_release(struct percpu_ref *ref)
@@ -93,13 +93,14 @@ net_devmem_alloc_dmabuf(struct net_devmem_dmabuf_binding *binding)
 	ssize_t offset;
 	ssize_t index;

-	dma_addr = gen_pool_alloc_owner(binding->chunk_pool, PAGE_SIZE,
+	dma_addr = gen_pool_alloc_owner(binding->chunk_pool,
+					1UL << binding->niov_shift,
 					(void **)&owner);
 	if (!dma_addr)
 		return NULL;

 	offset = dma_addr - owner->base_dma_addr;
-	index = offset / PAGE_SIZE;
+	index = offset >> binding->niov_shift;
 	niov = &owner->area.niovs[index];

 	niov->desc.pp_magic = 0;
@@ -113,12 +114,13 @@ void net_devmem_free_dmabuf(struct net_iov *niov)
 {
 	struct net_devmem_dmabuf_binding *binding = net_devmem_iov_binding(niov);
 	unsigned long dma_addr = net_devmem_get_dma_addr(niov);
+	size_t niov_size = 1UL << binding->niov_shift;

 	if (WARN_ON(!gen_pool_has_addr(binding->chunk_pool, dma_addr,
-				       PAGE_SIZE)))
+				       niov_size)))
 		return;

-	gen_pool_free(binding->chunk_pool, dma_addr, PAGE_SIZE);
+	gen_pool_free(binding->chunk_pool, dma_addr, niov_size);
 }

 void net_devmem_unbind_dmabuf(struct net_devmem_dmabuf_binding *binding)
@@ -163,6 +165,9 @@ int net_devmem_bind_dmabuf_to_queue(struct net_device *dev, u32 rxq_idx,
 	u32 xa_idx;
 	int err;

+	if (binding->niov_shift != PAGE_SHIFT)
+		mp_params.rx_page_size = 1U << binding->niov_shift;
+
 	err = netif_mp_open_rxq(dev, rxq_idx, &mp_params, extack);
 	if (err)
 		return err;
@@ -184,10 +189,12 @@ struct net_devmem_dmabuf_binding *
 net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,
 		       struct device *dma_dev,
 		       enum dma_data_direction direction,
-		       unsigned int dmabuf_fd, struct netdev_nl_sock *priv,
+		       unsigned int dmabuf_fd, unsigned int niov_shift,
+		       struct netdev_nl_sock *priv,
 		       struct netlink_ext_ack *extack)
 {
 	struct net_devmem_dmabuf_binding *binding;
+	size_t niov_size = 1UL << niov_shift;
 	static u32 id_alloc_next;
 	struct scatterlist *sg;
 	struct dma_buf *dmabuf;
@@ -213,6 +220,7 @@ net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,

 	binding->dev = dev;
 	binding->vdev = vdev;
+	binding->niov_shift = niov_shift;
 	xa_init_flags(&binding->bound_rxqs, XA_FLAGS_ALLOC);

 	err = percpu_ref_init(&binding->ref,
@@ -255,11 +263,7 @@ net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,
 		}
 	}

-	/* For simplicity we expect to make PAGE_SIZE allocations, but the
-	 * binding can be much more flexible than that. We may be able to
-	 * allocate MTU sized chunks here. Leave that for future work...
-	 */
-	binding->chunk_pool = gen_pool_create(PAGE_SHIFT,
+	binding->chunk_pool = gen_pool_create(niov_shift,
 					      dev_to_node(&dev->dev));
 	if (!binding->chunk_pool) {
 		err = -ENOMEM;
@@ -273,9 +277,12 @@ net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,
 		size_t len = sg_dma_len(sg);
 		struct net_iov *niov;

-		if (!IS_ALIGNED(len, PAGE_SIZE)) {
+		if (!IS_ALIGNED(dma_addr, niov_size) ||
+		    !IS_ALIGNED(len, niov_size)) {
 			err = -EINVAL;
-			NL_SET_ERR_MSG(extack, "dma-buf SG length must be PAGE_SIZE aligned");
+			NL_SET_ERR_MSG_FMT(extack,
+					   "dmabuf sg entry (addr=%pad, len=%zu) not aligned to niov size %zu",
+					   &dma_addr, len, niov_size);
 			goto err_free_chunks;
 		}

@@ -288,7 +295,7 @@ net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,

 		owner->area.base_virtual = virtual;
 		owner->base_dma_addr = dma_addr;
-		owner->area.num_niovs = len / PAGE_SIZE;
+		owner->area.num_niovs = len >> niov_shift;
 		owner->binding = binding;

 		err = gen_pool_add_owner(binding->chunk_pool, dma_addr,
@@ -454,7 +461,7 @@ int mp_dmabuf_devmem_init(struct page_pool *pool)
 	pool->dma_sync = false;
 	pool->dma_sync_for_cpu = false;

-	if (pool->p.order != 0)
+	if (pool->p.order != binding->niov_shift - PAGE_SHIFT)
 		return -E2BIG;

 	net_devmem_dmabuf_binding_get(binding);
--- a/net/core/devmem.h
+++ b/net/core/devmem.h
@@ -71,6 +71,8 @@ struct net_devmem_dmabuf_binding {
 	 */
 	struct net_iov **tx_vec;

+	unsigned int niov_shift;
+
 	struct work_struct unbind_w;
 };

@@ -93,7 +95,8 @@ struct net_devmem_dmabuf_binding *
 net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,
 		       struct device *dma_dev,
 		       enum dma_data_direction direction,
-		       unsigned int dmabuf_fd, struct netdev_nl_sock *priv,
+		       unsigned int dmabuf_fd, unsigned int niov_shift,
+		       struct netdev_nl_sock *priv,
 		       struct netlink_ext_ack *extack);
 struct net_devmem_dmabuf_binding *net_devmem_lookup_dmabuf(u32 id);
 void net_devmem_unbind_dmabuf(struct net_devmem_dmabuf_binding *binding);
@@ -122,10 +125,11 @@ static inline u32 net_devmem_iov_binding_id(const struct net_iov *niov)

 static inline unsigned long net_iov_virtual_addr(const struct net_iov *niov)
 {
-	struct net_iov_area *owner = net_iov_owner(niov);
+	struct dmabuf_genpool_chunk_owner *co =
+		net_devmem_iov_to_chunk_owner(niov);

-	return owner->base_virtual +
-	       ((unsigned long)net_iov_idx(niov) << PAGE_SHIFT);
+	return net_iov_owner(niov)->base_virtual +
+	       ((unsigned long)net_iov_idx(niov) << co->binding->niov_shift);
 }

 static inline bool
@@ -175,6 +179,7 @@ net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,
 		       struct device *dma_dev,
 		       enum dma_data_direction direction,
 		       unsigned int dmabuf_fd,
+		       unsigned int niov_shift,
 		       struct netdev_nl_sock *priv,
 		       struct netlink_ext_ack *extack)
 {
--- a/net/core/netdev-genl-gen.c
+++ b/net/core/netdev-genl-gen.c
@@ -11,6 +11,7 @@

 #include <uapi/linux/netdev.h>
 #include <net/netdev_netlink.h>
+#include <asm/page.h>

 /* Integer value ranges */
 static const struct netlink_range_validation netdev_a_page_pool_id_range = {
@@ -27,6 +28,11 @@ static const struct netlink_range_validation netdev_a_napi_defer_hard_irqs_range
 	.max	= S32_MAX,
 };

+static const struct netlink_range_validation netdev_a_dmabuf_rx_page_size_range = {
+	.min	= PAGE_SIZE,
+	.max	= U32_MAX,
+};
+
 /* Common nested types */
 const struct nla_policy netdev_lease_nl_policy[NETDEV_A_LEASE_NETNS_ID + 1] = {
 	[NETDEV_A_LEASE_IFINDEX] = NLA_POLICY_MIN(NLA_U32, 1),
@@ -106,10 +112,11 @@ static const struct nla_policy netdev_qstats_get_nl_policy[NETDEV_A_QSTATS_SCOPE
 };

 /* NETDEV_CMD_BIND_RX - do */
-static const struct nla_policy netdev_bind_rx_nl_policy[NETDEV_A_DMABUF_FD + 1] = {
+static const struct nla_policy netdev_bind_rx_nl_policy[NETDEV_A_DMABUF_RX_PAGE_SIZE + 1] = {
 	[NETDEV_A_DMABUF_IFINDEX] = NLA_POLICY_MIN(NLA_U32, 1),
 	[NETDEV_A_DMABUF_FD] = { .type = NLA_U32, },
 	[NETDEV_A_DMABUF_QUEUES] = NLA_POLICY_NESTED(netdev_queue_id_nl_policy),
+	[NETDEV_A_DMABUF_RX_PAGE_SIZE] = NLA_POLICY_FULL_RANGE(NLA_U32, &netdev_a_dmabuf_rx_page_size_range),
 };

 /* NETDEV_CMD_NAPI_SET - do */
@@ -219,7 +226,7 @@ static const struct genl_split_ops netdev_nl_ops[] = {
 		.cmd		= NETDEV_CMD_BIND_RX,
 		.doit		= netdev_nl_bind_rx_doit,
 		.policy		= netdev_bind_rx_nl_policy,
-		.maxattr	= NETDEV_A_DMABUF_FD,
+		.maxattr	= NETDEV_A_DMABUF_RX_PAGE_SIZE,
 		.flags		= GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DO,
 	},
 	{
--- a/net/core/netdev-genl-gen.h
+++ b/net/core/netdev-genl-gen.h
@@ -12,6 +12,7 @@

 #include <uapi/linux/netdev.h>
 #include <net/netdev_netlink.h>
+#include <asm/page.h>

 /* Common nested types */
 extern const struct nla_policy netdev_lease_nl_policy[NETDEV_A_LEASE_NETNS_ID + 1];
--- a/net/core/netdev-genl.c
+++ b/net/core/netdev-genl.c
@@ -1013,6 +1013,7 @@ netdev_nl_get_dma_dev(struct net_device *netdev, unsigned long *rxq_bitmap,
 int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info)
 {
 	struct net_devmem_dmabuf_binding *binding;
+	unsigned int niov_shift = PAGE_SHIFT;
 	u32 ifindex, dmabuf_fd, rxq_idx;
 	struct netdev_nl_sock *priv;
 	struct net_device *netdev;
@@ -1030,6 +1031,18 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info)
 	ifindex = nla_get_u32(info->attrs[NETDEV_A_DEV_IFINDEX]);
 	dmabuf_fd = nla_get_u32(info->attrs[NETDEV_A_DMABUF_FD]);

+	if (info->attrs[NETDEV_A_DMABUF_RX_PAGE_SIZE]) {
+		u32 rx_page_size = nla_get_u32(info->attrs[NETDEV_A_DMABUF_RX_PAGE_SIZE]);
+
+		if (!is_power_of_2(rx_page_size)) {
+			NL_SET_ERR_MSG_ATTR(info->extack,
+					    info->attrs[NETDEV_A_DMABUF_RX_PAGE_SIZE],
+					    "rx-page-size must be a power of 2");
+			return -EINVAL;
+		}
+		niov_shift = ilog2(rx_page_size);
+	}
+
 	priv = genl_sk_priv_get(&netdev_nl_family, NETLINK_CB(skb).sk);
 	if (IS_ERR(priv))
 		return PTR_ERR(priv);
@@ -1080,7 +1093,8 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info)
 	}

 	binding = net_devmem_bind_dmabuf(netdev, NULL, dma_dev, DMA_FROM_DEVICE,
-					 dmabuf_fd, priv, info->extack);
+					 dmabuf_fd, niov_shift, priv,
+					 info->extack);
 	if (IS_ERR(binding)) {
 		err = PTR_ERR(binding);
 		goto err_rxq_bitmap;
@@ -1221,7 +1235,7 @@ int netdev_nl_bind_tx_doit(struct sk_buff *skb, struct genl_info *info)
 	binding = net_devmem_bind_dmabuf(bind_dev,
 					 bind_dev != netdev ? netdev : NULL,
 					 dma_dev, DMA_TO_DEVICE, dmabuf_fd,
-					 priv, info->extack);
+					 PAGE_SHIFT, priv, info->extack);
 	if (IS_ERR(binding)) {
 		err = PTR_ERR(binding);
 		goto err_unlock_bind_dev;
--- a/tools/include/uapi/linux/netdev.h
+++ b/tools/include/uapi/linux/netdev.h
@@ -219,6 +219,7 @@ enum {
 	NETDEV_A_DMABUF_QUEUES,
 	NETDEV_A_DMABUF_FD,
 	NETDEV_A_DMABUF_ID,
+	NETDEV_A_DMABUF_RX_PAGE_SIZE,

 	__NETDEV_A_DMABUF_MAX,
 	NETDEV_A_DMABUF_MAX = (__NETDEV_A_DMABUF_MAX - 1)